[lxc-users] problem after migrating zfs pool to new disk

[Norberto Bensa]

Hello,

$ snap list
lxd     4.8        18324  latest/stable  canonical✓  -

I moved my zfs pool to a new disk and then renamed the pool.

Old name was zroot, new name: rz1-venkman-nvme.

The pool was migrated using zfs send|zfs receive.

While doing the migration, both pools were on the same machine.

The procedure was done with all the services down.

lxc shows:

$ lxc list
Error: Get "http://unix.socket/1.0": EOF

$ lxd
Error: This must be run as root

$ sudo lxd
[sudo] password for zoolook:
WARN[11-21|19:23:17] - Couldn't find the CGroup blkio.weight, disk priority will be ignored
EROR[11-21|19:23:26] Failed to start the daemon: Failed initializing storage pool "default": Failed to run: zpool import zroot: cannot import 'zroot': no such pool available
Error: Failed initializing storage pool "default": Failed to run: zpool import zroot: cannot import 'zroot': no such pool available

I figured I could run snap stop lxd, update the database at /var/snap/lxd/common/lxd/database/global/db.bin

I updated the value of the table storage_pools_config  to the correct volume, but the problem persist, but it is different

$ lxc list
Error: Get "http://unix.socket/1.0": dial unix /var/snap/lxd/common/lxd/unix.socket: connect: connection refused

$ sudo lxd
[sudo] password for zoolook:  
WARN[11-21|20:18:14]  - Couldn't find the CGroup blkio.weight, disk priority will be ignored  
EROR[11-21|20:18:15] Failed to start the daemon: Failed to start dqlite server: raft_start(): io: load closed segment 0000000000078237-0000000000078237: entries batch 2 starting at byte 48: entries count in preamb
le is zero  
Error: Failed to start dqlite server: raft_start(): io: load closed segment 0000000000078237-0000000000078237: entries batch 2 starting at byte 48: entries count in preamble is zero

Also. I don't know if this is related but there's at least two files I cannot read:

# cd /var/snap/lxd/common/ns
# ls -l
total 0
-r--r--r-- 1 root root 0 nov 21 19:48 mntns
-r--r--r-- 1 root root 0 nov 21 19:48 shmounts

# cat mntns  
cat: mntns: Invalid argument

# cat shmounts  
cat: shmounts: Invalid argument

Thanks for any help

Norberto

[Norberto Bensa]

I cloned the github repository and found docs/database.md

There's a patch.global.sql that did the trick :-)